Gene expression profiles as predictors of poor outcomes in stage II colorectal cancer: A systematic review and meta-analysis.

PURPOSE

The use of adjuvant therapy in stage II colorectal cancer (CRC) remains controversial. There is a need to identify more effective predictors than the traditional staging system to aid therapeutic decision-making. We performed a systematic review and meta-analysis of gene expression profiles (GEPs) to assess their utility for risk stratification and prediction of poor outcomes in stage II CRC.

PATIENTS AND METHODS

We performed a comprehensive literature search through December 2007. Studies were included if they reported GEP-based assays in patients with stage II CRC, and either subsequent cancer recurrence or death within 3 years. The prognostic likelihood ratio (LR) and odds ratio (OR) were calculated with 95% confidence intervals and pooled using the fixed-effects method. The weighted average sensitivity, specificity, and accuracy were also reported.

RESULTS

Eight cohorts involving 271 patients contributed to the analysis. The average accuracy, sensitivity, and specificity were 81.9%, 76.2%, and 84.5%, respectively, with a prognostic LR of 4.7 (95% CI, 3.2-6.8) and a prognostic OR of 15.1 (95% CI, 7.9-28.6). No evidence for significant interstudy heterogeneity was noted in either analysis. Subgroup analysis found no difference in results for the prediction of cancer recurrence or death.

CONCLUSION

This analysis demonstrates the promising potential of using GEP assays as predictors of poor outcomes in stage II CRC, such as cancer recurrence or death. To maximize their utility and availability, further studies will be needed to identify and validate specific gene signatures for poor prognosis in stage II CRC.